Abstract
In musculoskeletal medicine, with its different schools of approach, examination and therapy, diagnostic routines for low back pain (LBP) are characterised by non-reproducible and/or non-validated test procedures. Numerous diagnostic tests and procedures have been developed, sometimes for the same anatomical structure. This paper reviews the current status of the diagnostic process in relation to medical history, inspection, palpation, examination routines and tests, composed from the literature and from data collected by the International Federation for Manual/musculoskeletal Medicine. The results show a meagre picture with mainly low but sometimes acceptable kappa values Methodological flaws are one of the reasons for these poor results. A second major reason to explain the poor reproducibility of many tests is their absence of validity. To test the validity of a test a gold standard is needed for comparison, and this is usually lacking. A third and probably the most important reason is the fact that patients with LBP are a heterogeneous group with different medical histories, different clinical features, different aetiology and prognosis. A valid classification system for LBP is needed.